Computing system for determining resource to perform operation on data and operating method thereof

Abstract

The present disclosure relates to a computing system. The computing system may include a memory system including a plurality of memory devices configured to store raw data and a near data processor (NDP) configured to receive the raw data by a first bandwidth from the plurality of memory devices and generate intermediate data by performing a first operation on the raw data, and a host device coupled to the memory system by a second bandwidth and determining a resource to perform a second operation on the intermediate data based on a bandwidth ratio and a data size ratio.

What is claimed is: 
     
         1 . A memory system comprising:
 a plurality of memory devices configured to store raw data; and   at least one processor configured to:   receive the raw data by a first bandwidth from the plurality of memory devices,   receive a request to process the raw data by a second bandwidth from a host, and   transmit, in response to the request, one of intermediate data generated by performing a first operation on the raw data or final data generated by performing a second operation on the intermediate data to the host based on a bandwidth ratio and a data size ratio,   wherein the bandwidth ratio is a ratio of the second bandwidth to the first bandwidth, and   wherein the data size ratio is a ratio of a size of the final data to a size of the raw data.   
     
     
         2 . The memory system of  claim 1 , wherein when the bandwidth ratio is greater than the data size ratio, the at least one processor generates the intermediate data and transmit the intermediate data to the host. 
     
     
         3 . The memory system of  claim 2 , wherein the final data is generated by performing the second operation on the intermediate data by the host. 
     
     
         4 . The memory system of  claim 1 , wherein when the bandwidth ratio is smaller than or equal to the data size ratio, the at least one processor generates the final data and transmits the final data to the host. 
     
     
         5 . The memory system of  claim 1 , wherein the at least one processor is further configured to receive a notification indicating that a reception of the intermediate data or the final data has been completed from the host. 
     
     
         6 . The memory system of  claim 1 , wherein the request is generated by the host based on a result of comparing the bandwidth ratio with the data size ratio. 
     
     
         7 . The memory system of  claim 1 , wherein the at least one processor comprises:
 a memory controller configured to control the plurality of memory devices to perform a memory operation including a read operation, an erase operation, and a program operation; and   at least one processor core configured to perform a processing operation including the first operation and the second operation.   
     
     
         8 . The memory system of  claim 7 , wherein the processing operation is configured as at least one or more of a pulling operation, a vector multiplication, a vector sum and a matrix multiplication. 
     
     
         9 . A method of operating a memory system including a plurality of memory devices, the method comprising:
 receiving a raw data by a first bandwidth from the plurality of memory devices;   receiving a request to process the raw data by a second bandwidth from a host; and   transmitting, in response to the request, one of intermediate data generated by performing a first operation on the raw data or final data generated by performing a second operation on the intermediate data to the host based on a bandwidth ratio and a data size ratio,   wherein the bandwidth ratio is a ratio of the second bandwidth to the first bandwidth, and   wherein the data size ratio is a ratio of a size of the final data to a size of the raw data.   
     
     
         10 . The method of  claim 9 , wherein the transmitting the one of the intermediate data or the final data comprises, in response to the bandwidth ratio is greater than the data size ratio, generating the intermediate data and transmit the intermediate data to the host. 
     
     
         11 . The method of  claim 10 , wherein the final data is generated by performing the second operation on the intermediate data by the host. 
     
     
         12 . The method of  claim 9 , wherein the transmitting the one of the intermediate data or the final data comprises, in response to the bandwidth ratio is smaller than or equal to the data size ratio, generating the final data and transmits the final data to the host. 
     
     
         13 . The method of  claim 9 , further comprising receiving a notification indicating that a reception of the intermediate data or the final data has been completed from the host. 
     
     
         14 . The method of  claim 9 , wherein the request is generated by the host based on a result of comparing the bandwidth ratio with the data size ratio. 
     
     
         15 . The method of  claim 9 , wherein each of the first operation and the second operation is configured as at least one or more of a pulling operation, a vector multiplication, a vector sum and a matrix multiplication.